# One Admires, from Victor Niederhoffer

April 28, 2016 |

One admires the typical high tech earnings report. Made 15 cents excluding certain items. Those items include salaries. With salaries loss of 20 cents. Twitter an example today. Something rotten in the way sale growth of 30% versus estimate 35% is cause for the 15% decline rather than they can't earn money.

## Ed Stewart writes:

If you think that is bad you should read the "value" stock releases. Every quarter it is "but for" earnings and the analysts play along with the BS to the extent that the fake numbers are the ones that are the default in factset.

Name

Email

Website

1. Yucheng Pan on April 28, 2016 5:20 pm

Fallacies of Sharpe Ratio and the new Victor Ratio

I never bothered using Sharpe Ratio (excess return divided by standard deviation) until today and its fallacies are immediately obvious. For example, a trading strategy with random yearly return between 0% and 100% (normally distributed) has a Sharpe Ratio of 1.7x (based on 10000 years), but all normally distributed random returns between 0% and any positive number have Sharpe Ratio of 1.7x (based on 10000 years) . This is clearly wrong as yearly return between 0 & 0.1% (average 0.05%) is clearly different from 0 & 1000000% (average 500000%). What is more fallacious is that any one large-enough outsized gain for a year would drop the Sharpe Ratio to zero from 1.7x for all the above experiments.

Here is a more accurate way to measure the performance of a trading strategy: ER * (1 – sgn * ED)^2, where ER is excess return, ED is maximum excess drawdown, and sgn is either 1 for positive ER or -1 for negative ER. I would like to name this ratio as Victor Ratio in honor of our Chair and our desire for victory. For example, if benchmark SPY has 10% return and 30% maximum drawdown, trading strategies with 50% return and maximum drawdowns of 5%, 30%, and 90% would have Victor Ratio of 0.6, 0.4, and 0.06, respectively (to replace the 50% return with -50%, the Victor Ratio would be -0.3, -0.6, and -1.5, respectively).

2. Michael Kelley on April 30, 2016 1:18 am

Yes, but twitter never sleeps. Neither does FB or GOOG or AAPL or et alia. Earnings here do not matter because the reason the universe spawned these symbols is to produce a change about which we can only theorize and imagine. Let them lose as much money as they can and their losses will not invalidate their role within the cycle. If not TWTR, then some other 4 letter name will emerge.

The wave is more important than the breaks upon the rocks.

Full Disclosure? I own none of these shares and will not until they are hat sizes once again.